| |
In lugnet.loc.jp, Paul Davidson writes:
> Lugnet seems to have serious problems with non-latin encoding --
Hmm, I'm not so sure.
> I'm not even using 8-bit encoding methods.
Hmm, well, but you were using Microshaft Look-Out to post the messages. The
funny thing is, all of the non-Latin characters were converted (somewhere)
to ?'s. Most likely, methinks, this happened not at the newsserver but at
the client end. AFAIK, the guts of newsserver don't ever munge the body
content of a message based on the low-level characters encodings. So regular
JIS should in theory work, and Shift-JIS should definitely work.
One thing for sure, though -- even if the raw NNTP article comes in OK, it's
going to be broken when it's served back via HTTP -- because the current
display library which kicks out webpages from NNTP articles converts anything
above 128 to HTML entities like &foo; and &#nnn;. It wouldn't be tough,
though, to have it emit raw 8-bit J-text if the headers were set properly
in the article.
> Maybe Todd can look into it?
Sure -- if you can e-mail me an attachment with a small snippet of ShiftJIS-
encoded text, I can try a bunch of things to .off-topic.test and look a
what's being received and what, if anything, is going wrong if what's being
received is right.
Can you make a block each of:
- Hiragana
- Katakana
- Kanji
and each of those as
- JIS
- Shift-JIS
If so, that would be great. But if not, I can probably dig something like
that up from somewhere on the net.
--Todd
[followups to .admin.general]
|
|
Message is in Reply To:
 | | conclusion
|
| Lugnet seems to have serious problems with non-latin encoding -- I'm not even using 8-bit encoding methods. Maybe Todd can look into it? -- Paul Davidson Paul Davidson <tinman@direct.ca> wrote in message news:Fou2sK.AsE@lugnet.com... (...) (25 years ago, 24-Jan-00, to lugnet.loc.jp)
|
8 Messages in This Thread:
  
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
